Parsing Coordination for Spoken Language Understanding